3
$\begingroup$

I could really use some advice with an interaction interpretation. I have conducted a 3 way repeated measures anova with terms A, B and C. A has 2 levels, B has 3 levels and C has 4 levels.

My results show a significant main effect of A and a significant interaction effect of BxC. I am aware that if there is an interaction effect then one should not interpret the main effect.

Does that hold true in this situation when the variable A with a main effect does not experience any significant interactions?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
3
$\begingroup$

The general advice to not interpret a main effect if there are interactions is too strong, at the very least. For example, if the variable of interest necessarily has a limited range of values (e.g. sex), and there is a strong main effect, then it's perfectly reasonable to interpret the main effect even if there is also an interaction.

Regardless, the advice only applies to the variables for which there are interactions. So, in your case interpret your A main effect and your B x C interaction separately.

Improve this answer
$\endgroup$
1
  • $\begingroup$ @user28327 If that answers your question, you could “accept” the answer by clicking the little check mark under the question's score on the left. $\endgroup$
    – Gala
    Aug 7, 2013 at 13:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.